Youth aggression (e.g., bullying, fighting) and serious violence (e.g., shootings, robbery) places a considerable emotional and financial burden on society, making effective prevention a paramount public health concern. While several empirically supported prevention programs have been developed to target elementary school children at risk for exhibiting chronic aggression and serious violence (A-SV) into adolescence, no screening tools have been explicitly designed to identify children in the community who should be enrolled in these programs. As it stands, children are recruited into these interventions based on various ad hoc screening procedures designed to delineate youth with early conduct problems (e.g., defiance, lying, fighting). Recent longitudinal examinations of these screening strategies indicate that they are relatively poor at accurately identifying those children who will exhibit a persistent pattern of A-SV across adolescence. As a result, costly prevention programs are being unnecessarily administered to low-risk children, while a sizable portion of children most in need of violence prevention services are overlooked. To address this issue, the proposed study will use a multi-staged measurement development strategy to create a set of informant-specific (i.e., parent, teacher, youth report) screening instruments explicitly designed to identify children at risk for exhibiting chronic A-SV across adolescence. First, secondary data analysis of four large longitudinal studies will be used to identify the risk factor across multiple domains (e.g., low school bonding, peer deviance) that most accurately and consistently delineate elementary school children at risk for exhibiting A-SV across multiple years during adolescence (~ages 13-20). This will include identifying informant-specific risk factors that help to optimize classification accuracy. Next, state-of-the-art qualitative and quantitative measurement development techniques will be used to create three informant-specific screening instruments that assess the risk factors that best classify children at risk for exhibiting chronic A-SV across the longitudinal studies. A comprehensive item bank designed to assess each risk factor will be created by systematically identifying and extracting items from all relevant instruments that exist in the literature. A Consortium of international experts on youth violence will systematically refine the item set to optimize construct coverage and content clarity, and cognitive interviews with targeted informants will be used to further identify and rectify problematic items. A national sample of 2520 elementary school children will then be recruited to psychometrically refine and validate the violence risk screening tools. Using item response theory and computer adaptive testing methods, an efficient subset of items that minimize measurement error and optimize precision will be selected for inclusion in the final version of the screening instruments. The deliverables from this project will be a set of free, empirically- based, and precise risk screening instruments that can be used in multiple settings to identify children most in need of targeted violence prevention services.